“Data-Driven Thinking” is written by members of the media community and contains fresh ideas on the digital revolution in media.
Today’s column is written by Dillon Roulet, founder and CEO at Syndic8.
Advertising without an audience sounds like an oxymoron, but it’s a reality that’s destined to take hold sooner, rather than later.
When programmatic came along, it upended the entire foundation on which digital advertising was created. In the old days, having a solid digital strategy meant buying impressions in bulk to cast a large net across the web. Targeting was limited (at best) and personalization was trivial.
Times have changed, with advertisers shifting more and more of their spend into programmatic. And while personalization has only just begun to come to fruition, just as we no longer buy impressions in bulk, we won’t always buy audience segments en masse.
Just like our fingerprints, no two personalities are identical. As we head into the twilight of old-school display, the only solution moving forward is to focus on the individual – not the audience.
The Audience Model Doesn’t Account For Emotional Variation
Currently, much of ad tech is focused on discovering new audience segments. We’re consumed with connecting dozens of variables to predict and gauge user intent. The reason? Right now, it’s the scalable option.
Unfortunately, this doesn’t take into consideration how user intent is dependent on a variety of complex emotional variables. An individual may be young, affluent and interested in golf, but if they’re agitated about a suddenly tightened budget, they’re not going after that new putter you’re trying to sell.
So many emotional factors influence user intent that we fail to take into account when running campaigns. Grief, anxiety, excitement, fluster, confusion – the list goes on. Emotional variation must be understood by our technology to deliver more effective impressions.
Undoubtedly, we’ll need intelligent machinery to do the heavy lifting. Understanding and predicting what content appeals to a user at that specific point in time requires technology focused on getting to know users at an individualized level – rather than categorizing them into pre-molded audience segments.
Intelligent Data Management Fuels Hyperpersonal Experiences
Obviously, to do this requires a heaping pile of data per user. Data management platforms (DMPs) will undoubtedly need to evolve for us to serve such hyperpersonalized campaigns, as well as widespread adaptation of universal IDs.
But the potential for bots to manage and mesh troves of user data into the right nooks and crannies is in our line of sight.
DMP giants have already aggressively responded to this demand. Salesforce’s Einstein and Oracle’s Adaptive Intelligent Apps have emitted clear signals that future data management will be smarter and more malleable. Being smarter, malleable and agile means that intelligence-powered platforms respond to cluttered or organized data in the same fashion. To top-notch AI, there is no differentiation and never too much data to handle.
Measurement: Moving Away From Quantity, Pushing For Quality
We’re finally seeing the end of times for older quantitative metrics and focal shift toward insight-rich KPIs. In a world where quality trumps quantity, it’s critical we discover how each ad impacts an individual user. Did the experience skew positive or negative? To what degree was the user influenced to make a purchasing decision? How does this compare to other campaigns viewed by this individual?
These are complex questions that require advanced answers. But the results will paint a much more holistic picture of the effectiveness of specific campaigns. Looking ahead, we need to place greater emphasis on user interaction, especially as it relates to video and native content.
Google’s launch of its time-spent metric last year proved demand for adaptive measurement is on the ascent. And while we should undoubtedly be collecting the time users spend on content, individual variation is too diverse for this metric to end up in the industry’s KPI repertoire. Hyperpersonalization requires a more spectrum-centric measurement approach, as opposed to the two-dimensional methods we use today.
When content becomes personalized, so should measurement. Luckily, just as AI-powered advertising can learn and evolve on an individual basis, so can the metrics we use to diagnose individual user experiences.
If we’re truly committed to combatting display fatigue, we need to upend the entire model on which campaigns reach and affect the end user. Personalized campaigns require personalized strategy. And when we focus on the individual, rather than the group, everyone in the audience has our attention.
Follow AdExchanger (@adexchanger) on Twitter.
